Download Exposure to excess glucocorticoids alters dendritic

Brain Research, 531 (1990) 225-231
Elsevier
225
BRES 15975
Exposure to excess glucocorticoids alters dendritic morphology of
adult hippocampal pyramidal neurons
Catherine S. Woolley, Elizabeth Gould and Bruce S. McEwen
Laboratory of Neuroendocrinology, The Rockefeller University, NY IO021 (U.S.A.)
(Accepted 1 May 1990)
Key words: Hippocampus; Pyramidal cell; Granule cell; Glucocorticoid; Golgi impregnation; Dendrite; Morphology
We have used Golgi-impregnated tissue to demonstrate that exposure to excess glucocorticoids alters dendritic morphology in a specific
population of neurons in the adult rat hippocampus. Daily injection of 10 mg of corticosterone for 21 days resulted in decreased numbers of
apical dendritic branch points and decreased total apical dendritic length measured in a 100-gin-thick section in CA3 pyramidal cells compared
to sham-injected and non-injected controls. In contrast, no changes were observed in CA3 pyramidal cell basal dendritic morphology.
Furthermore, no changes were observed in the dendritic morphology of CA1 pyramidal cells or granule cells of the dentate gyrus.
Cross-sectional cell body area of any of the 3 cell types examined in this study was unaffected by cortieosterone treatment. Finally, qualitative
analysis of Nissl-stained tissue from the same brains revealed increased numbers of darkly staining, apparently shrunken CA3 pyramidal cells
in corticosterone treated compared to control brains. The changes in dendritic morphology we have observed may be indicative of neurons
in the early stages of degeneration, as prolonged exposure to high levels of corticosterone has been shown by others 29 to result in a loss of
CA3 pyramidal cells. Additionally, these results suggest possible structural alterations which may occur under physiological conditions in which
corticosterone levels are chronically elevated such as in aged animals.
INTRODUCTION
As early as 1969, Aus Der Muhlen and Ockenfels 2
demonstrated that elevated levels of glucocorticoids, the
adrenal steroids secreted in response to stress 26-28, are
toxic to neurons in the adult guinea pig hippocampus.
Since then, numerous studies have sought to understand
this neurotoxic p h e n o m e n o n and its implications for
hippocampal function under physiological conditions in
which glucocorticoids levels are elevated, e.g. in aged 19'
26,40 or stressed 26-28 animals. Landfield et al. 2° have
demonstrated a role for glucocorticoids in the progressive
loss of hippocampal neurons with age in the rat as
mid-life adrenalectomy attenuates the decrease in hippocampal neuron number normally observed in aged
animals. In addition, U n o et al. have found that
hippocampal neuronal loss occurs in vervet monkeys
subjected to severe chronic social stress 41.
The damaging effects of chronic glucocorticoid exposure in the hippocampus are of particular interest as this
neural structure has been proposed to be involved in
cognitive processes such as learning and memory 23'38 as
well as in aspects of neuroendocrine control 5'1s. As
elevated glucocorticoid levels have been implicated in
hippocampal n e u r o n l o s s 20'29 and such hippocampal
neuron loss has been linked to both cognitive 2° and
neuroendocrine 26'28 impairments in aged animals, a
better understanding of glucocorticoid-induced neuronal
damage may shed light on the processes which contribute
to alterations in brain function observed with aging x°'26"28
and chronic stress 6.
The cellular mechanisms underlying the effects of
chronic glucocorticoid exposure in the hippocampus are
not well understood. While it seems likely that glucocorticoid-induced neuronal damage stems from a variety of
biochemical and/or other factors, alterations in neuronal
morphology may also be involved in the response of the
brain to chronically elevated glucocorticoid levels. A
number of studies have suggested neuronal morphology,
particularly dendritic architecture, to be a modulator of
neuronal physiology and thereby to influence brain
function x5'24. As adult hippocampal neurons have been
shown to be morphologically sensitive to low glucocorticoid levels induced by adrenalectomy 11, it seems quite
possible that elevated levels of glucocorticoids may also
result in an altered neuronal morphological profile. As
such, excess glucocorticoids may exert negative effects on
hippocampal function not only by reducing the absolute
number of hippocampal neurons, but by altering the
morphology of surviving neurons as well. To date, no
Correspondence: C.S. Woolley, Laboratory of Neuroendocrinology, The Rockefeller University, 1230 York Avenue, New York, NY 10021,
U.S.A.
0006-8993/90/$03.50 (~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
226
study has e x a m i n e d the effects of c h r o n i c g l u c o c o r t i c o i d
TABLE 1
a d m i n i s t r a t i o n on the m o r p h o l o g y o f h i p p o c a m p a l neu-
Number of dendritic branch points present in a lO0-pm-thick section
rons. In o r d e r to d e t e r m i n e w h e t h e r excess glucocorticoids alter the structure of h i p p o c a m p a l n e u r o n s , we
h a v e p e r f o r m e d a q u a n t i t a t i v e m o r p h o l o g i c a l analysis of
Values represent mean + S.E.M. n refers to the number of animals in
each treatment group. Data were analyzed with one-way analysis
followed by Tukey HSD posthoc comparisons.
the d e n d r i t i c a r c h i t e c t u r e of G o l g i - i m p r e g n a t e d h i p p o c a m p a l C A 3 p y r a m i d a l cells, C A 1 p y r a m i d a l cells and
g r a n u l e cells of the d e n t a t e gyrus f r o m c o r t i c o s t e r o n e injected,
sham-injected
and n o n - i n j e c t e d
c o n t r o l ani-
mals. Q u a l i t a t i v e a s s e s s m e n t of Nissl-stained tissue was
also p e r f o r m e d .
MATERIALS AND METHODS
Animals treatonents and histological procedures
Male Sprague-Dawley (Charles River) rats (250-300 g) were
subjected to one of the following treatments: (1) subcutaneous
injection of 10 mg corticosterone in 250/A sesame oil once daily for
21 days; or (2) subcutaneous injection of the vehicle alone once
daily for 21 days. The remaining rats did not receive any treatments
for the duration of the experiment. The dose of corticosterone used
in this study had previously been shown to result in serum
corticosterone levels which are initially above the physiological
range and then decline to basal levels within 24 h ~z.
At the end of the treatment period, these rats were deeply
anesthetized with Ketamine HCI and transeardially perfused with
150 ml 4.0% paraformaldehyde in 0.1 M phosphate buffer with
1.5% (v/v) picric acid. Brains were postfixed in the perfusate for 5
days. Sections, 100/~m thick, were cut with a Vibratome into a bath
of 3.0% potassium dichromate in distilled water. Some sections
were rinsed in 0.1 M phosphate buffer, mounted onto gelatinized
slides and stained for Nissl with Cresyl violet. The remaining
sections were then processed according to a modified version of the
single-section Golgl impregnation procedure 9. Briefly, the brain
sections were incubated in 3.0% potassium dichromate in distilled
water overnight. The sections were then rinsed in distilled water,
mounted onto plain slides and a coverslip was glued over the
sections at the 4 corners. These slide assembfies were incubated in
1.5% silver nitrate in distilled water overnight in the dark. Following
this, the slide assemblies were dismantled, the tissue sections rinsed
in distilled water and then dehydrated in 95% followed by absolute
ethanol. The sections were then cleared in Histoclear, mounted onto
gelatinized slides and coverslipped under Permount.
Data analysis
Slides containing brain sections were coded prior to quantitative
analysis; the code was not broken until the analysis was complete.
In order to be selected for analysis, Golgl-impregnated neurons had
to possess the following characteristics: (1) location in the appropriate subregion of the rostral hippocampns; (2) dark and consistent
impregnation throughout the extent of all of the dendrites; (3)
relative isolation from neighboring impregnated cells which could
interfere with analysis; and (4) a cell body in the middle third of the
tissue section in order to avoid analysis of impregnated neurons
which extended largely into other sections. For each brain, 5 CA3
pyramidal cells (3 from the CA3c region and 2 from the CA3b
region), 5 CA1 pyramidal cells and 5 dentate gyrns granule cells
from the suprapyramidal blade (3 with a single primary dendrite and
2 with multiple primary dendrites) were selected. Each selected
neuron was drawn at 500x using a camera lucida drawing tube.
From these drawings, the number of dendritic branch points within
a 100-/~m-thick section of each dendritic tree was determined for
each selected neuron. In addition, the length of the dendrites
present in a 100-/~m-thick section was determined for each dendritic
tree using the SMI (Southern Micro Instruments) image analysis
morphometry program. Cross-sectional cell body area measure-
Neural region
Non-injected Sham
CA3 pyramidal cell apical
dendrites
12.6 + 1.1
(n = 6)
CA3 pyramidal cell basal
dendrites
9.8 + 0.9
(n = 6)
CA1 pyramidal cell apical
dendrites
23.4 + 1.5
(n = 4)
CA1 pyramidal cell basal
dendrites
11.8 + 0.8
(n = 4)
Dentate gyrus granule
cell dendrites
6.8 + 0.6
(n = 4)
CORT
12.4 _+0.8
(n = 4)
7.3 +_0.5*
(n = 5)
11.0 + 1.5
(n = 4)
10.8 + 0.6
(n = 5)
27.5 + 2.0
(n = 4)
24.8 + 1.2
(n = 4)
13.9 + 1.0
(n = 4)
13.7 + 0.8
(n = 4)
7.2 + 0.2
(n = 4)
7.5 + 0.3
(n = 5)
* Significant difference from non-injected and sham (P < 0.007).
ments were also made for each of the 3 cell types. Ten cell bodies
per cell type per brain were traced at 1250x using a camera lucida
drawing tube and cross-sectional area was determined using the SMI
morphometry program. Means were determined for each variable
for each brain and the resulting values were subjected to a one-way
ANOVA with Tukey HSD posthoc comparisons.
RESULTS
A t t h e e n d of t h e t r e a t m e n t p e r i o d , all a n i m a l s in e a c h
treatment
group appeared
to b e h e a l t h y .
Qualitative
analysis of G o l g i - i m p r e g n a t e d tissue r e v e a l e d sufficient
n u m b e r s of C A 3 p y r a m i d a l cells, C A 1 p y r a m i d a l cells
and d e n t a t e gyrus g r a n u l e cells w h i c h m e t o u r s e l e c t i o n
criteria in brains f r o m e a c h t r e a t m e n t g r o u p (see T a b l e s
I, II and III). T h e s e brains w e r e s u b j e c t e d to q u a n t i t a t i v e
analysis.
Q u a n t i t a t i v e analysis o f G o l g i - i m p r e g n a t e d C A 3 pyramidal
cells r e v e a l e d
significant
differences
in the
n u m b e r of apical d e n d r i t i c b r a n c h p o i n t s p r e s e n t in a
100-/am-thick section w i t h c o r t i c o s t e r o n e ( C O R T ) treatm e n t (F2,12 = 10.9; P < 0.003). C A 3 p y r a m i d a l cells o f
CORT-treated
animals
dendritic branch
had
significantly f e w e r apical
points than CA3
pyramidal
cells of
e i t h e r s h a m - i n j e c t e d ( P < 0.007) o r n o n - i n j e c t e d ( P <
0.004)
control animals
(Figs.
1 and
2; T a b l e
I). In
contrast, n o d i f f e r e n c e s in t h e n u m b e r of C A 3 p y r a m i d a l
cell basal d e n d r i t i c b r a n c h p o i n t s w e r e d e t e c t e d with
C O R T t r e a t m e n t (Figs. 1 and 2; T a b l e I). Q u a n t i t a t i v e
analysis of C A 1 p y r a m i d a l cells r e v e a l e d n o significant
d i f f e r e n c e s b e t w e e n t r e a t m e n t g r o u p s in t h e n u m b e r of
d e n d r i t i c b r a n c h points p r e s e n t in a 100-#m-thick section
in e i t h e r t h e apical or basal d e n d r i t i c t r e e (Table I).
227
:iii:i:i:iiiii!iii!:
• i
~
i [i
i : ~!ii
R
..............
Fig. 1. Golgi-impregnated CA3c pyramidal cells from brains of sham-injected (A) and CORT-injected (B) animals. These neurons were
selected to approximate mean values for both numbers of dendritic branch points and dendritic length. Camera lucida drawings of these cells
are shown in Fig. 2. Scale bar = 50 g m and applies to both frames.
TABLE II
Total dendritic length present in a l O0-1zm-thicksection
Values represent mean 4- S.E.M. n refers to the number of animals in
each treatment group. Data were analyzed with one-way analysis of
variance followed by Tukey HSD post hoc comparisons.
Neuralregion
Non-injected
(#m)
CA3 pyramidal cell
apical dendrites
1435.6 4- 92.8
(n = 6)
CA3 pyramidal cell
basal dendrites
1069.2 + 67.7
(n = 6)
CA1 pyramidal cell
apical dendrites 2038.3 4- 162.9
(n = 4)
CA1 pyramidal cell
basaldendrites
1380.04- 154.6
(n = 4)
Dentate gyrus granule cell dendrites 1043.5 4- 76.4
(n = 4)
Sham
(1urn)
CORT
(gm)
Similarly, no significant differences with CORT treatment were detected in the number of dendritic branch
points present in a 100-#m-thick section in dentate gyrus
granule cells (Table I).
In addition to the observed changes in the number of
apical dendritic branch points in CA3 pyramidal cells,
T A B L E III
Cross-sectional cell body area
1607.4 + 228.0
(n = 4)
908.8 + 77.7*
(n = 5)
1265.6 + 226.6 1062.6 4- 29.9
(n = 4)
(n = 5)
Values represent mean +_ S.E.M. n refers to the number of animals in
each treatment group. Data were analyzed with one-way analysis of
variance followed by Tukey HSD post hoc comparisons. No
significant differences were detected.
Neuralregion
2784.2 __+289.5 2311.9 __+108.3
(n = 4)
(n = 4)
CA3 pyramidal cell
1667.64-64.1
(n = 4)
1647.4+67.0
(n = 4)
1276.7 4- 81.4
1279.8 _+96.8
(n = 4)
(n = 5)
* Significant difference from non-injected and sham (P < 0.05).
CA1 pyramidal cell
Dentate gyrus granule
cell
Non-injected
Sham
CORT
O,m~)
(~r? )
(~,d )
410.0 + 12.4
410.9 + 8.2
380.2 4- 11.1
(n = 6)
(n = 4)
(n -- 6)
225.0 4- 4.7
(n = 6)
238.0 4- 6.3
(n = 4)
228.6 4- 5.9
(n = 6)
176.6 + 3.8
182.8 + 3.1
171.1 4- 3.6
(n = 6)
(n = 4)
(n = 6)
228
J
Fig. 2. Camera lucida drawings of the same CA3c pyramidal cells
from the brains of sham-injected (A) and CORT-injected (B)
animals depicted in Fig. 1. For these cells, numbers of dendritic
branch points are: (A) apical, 15 and basal, 10; (B) apical, 9 and
basal, 9. Dendritic length values for these cells are: (A) apical,
1636.9/~m and basal, 1164.7 ktm; (B) apical, 1136.3/~m and basal,
1285.0/~m. Note the decrease in both number of dendritic branch
points and dendritic length in the apical tree of (B) compared to (A)
whereas there is no change in these parameters in the basal tree.
Scale bar = 50/tm and applies to both frames.
C O R T treatment resulted in significant differences in the
total apical dendritic length present in a lO0-ktm-thick
section in CA3 pyramidal cells (F2,12 = 7.5; P ~Q 0.009).
Total dendritic length present in a 100-/~m-thick section of
the apical dendritic tree in CA3 pyramidal cells of
CORT-treated animals was significantly reduced compared to either sham injected (P < 0.008) or non-injected
(P < 0.05) control animals (Figs. 1 and 2; Table II). In
contrast, no significant differences in total basal dendritic
length in a 100-~m-thick section in CA3 pyramidal cells
were observed with C O R T treatment (Figs. 1 and 2;
Table II). Additionally, no significant differences in
either total apical or basal dendritic length present in a
100-/~m-thick section of CA1 pyramidal cells were observed with C O R T treatment (Table I1). Likewise, in
dentate gyrus granule cells, no significant differences in
total dendritic length present in a 100-/~m-thick section
were detected between treatment groups (Table II).
No changes in CA3 pyramidal cell cross-sectional cell
body area were observed with corticosterone treatment
(Table III). Similarly, no differences in CA1 pyramidal
cell or dentate gyrus granule cell cross-sectional cell body
area were observed between treatment groups (Table
III).
Qualitative examination of Nissl-stained tissue revealed no obvious signs of degeneration in the CA3
pyramidal cell, CA1 pyramidal cell or dentate gyrus
granule cell layers in any of the treatment groups in this
study. Very few pyknotic cells (0-1/100-/~m-thick section)
were observed in the CA3 pyramidal cell layer of brains
from any treatment group; they were observed in brains
from both control and C O R T - t r e a t e d animals. However,
the presence of a small number of darkly staining,
apparently shrunken cells in the CA3 pyramidal cell layer
(Fig. 3) was noted in some brains from both C O R T treated and control animals. Such apparently unhealthy
pyramidal cells were observed in more CORT-treated
than control brains and within each brain in which they
were noted, they were seen more frequently in C O R T treated than control brains. In some cases, these cells
appeared elongated, staining well into the apical dendritic shaft. In other cases, while still possessing visible
dendrites, shrunken cells appeared small and irregularly
shaped. Such cells did not seem to be associated with
areas of degeneration as they were usually observed
adjacent t o apparently healthy CA3 pyramidal cells (Fig.
3). In some, but not all, C O R T - t r e a t e d brains, an
increase was noted in the number of very small, darkly
staining cells which appeared to be glia.
DISCUSSION
Fig. 3. Nissl-stained CA3c pyramidal cells in the brain of a
CORT-treated animal. Note the presence of several small, darkly
staining cells (arrowheads) adjacent to apparently healthy pyramidal
neurons (long arrows). Scale bar = 25 ,urn.
These results demonstrate that chronic administration
of high levels of corticosterone to the adult rat results in
alterations in dendritic morphology of Golgi-impregnated
229
hippocampal CA3 pyramidal cells. Specifically, we have
observed decreases in the number of apical dendritic
branch points and total apical dendritic length in a
100-/~m-thick section with C O R T treatment. The fact that
we observed no differences in cross-sectional cell body
area supports the notion that the cells selected for
quantitative analysis in this study were representative of
the same neuronal population between treatment groups.
The changes in dendritic morphology, i.e. the apparent atrophy of the apical dendritic tree, we have observed
are consistent with the previously reported finding that
prolonged exposure to excess CORT is damaging to
hippocampal pyramidal cells in the CA3 region 29. However, qualitative analysis of Nissl-stained tissue provided
no conclusive evidence for CORT-induced death of CA3
pyramidal neurons; in the CA3 region, no increase in the
number of pyknotic cells was observed in the hippocampi
of CORT-treated compared to control animals and the
small numbers of apparently unhealthy pyramidal cells
observed were detected in brains from each treatment
group, albeit substantially more frequently in CORTtreated than in control animals. However, an apparent
increase in the number of glial cells in the CA3 region of
CORT-treated brains could indicate neuronal degeneration as astrocytes are known to migrate toward and
proliferate within an area of neuronal damage 36.
Sapolsky et al. 29 have reported that chronic CORT
administration results in a shift toward decreasing values
in the distribution of cross-sectional cell body areas and
in a concomitant decrease in the absolute number of
pyramidal cells in the CA3 region. Although we do not
have evidence for such dramatic effects of CORT
treatment, there are differences between our study and
that of Sapolsky et al. 29 in both the age and strain of
animals used as well as the dose of CORT used and
duration of treatment. Thus it is quite possible that had
we extended the duration of our CORT treatment, we
might have obtained more convincing evidence for
neuronal loss. In addition, as our quantitative analysis
was based upon Golgi-impregnated tissue, it is possible
that our criteria for selection of neurons for analysis
excluded some degenerating neurons, i.e. perhaps degenerating neurons are less likely to be well impregnated.
Thus, the changes in dendritic morphology reported here
may be indicative of neurons in the very early stages of
degeneration. It should be noted that more subtle effects
of CORT can be detected relatively rapidly as increases
in the surface densities of rough endoplasmic reticulum
and Goigi apparatus in pyramidal cells of the CA3 and
CA1 regions as well as dentate gyrus granule cells have
been observed with as few as 3 days of CORT
treatment 21.
Sapolsky 3° has suggested that CORT acts directly on
hippocampal pyramidal neurons to effect a state of
increased metabolic vulnerability thereby compromising
their ability to survive subsequent metabolic challenges.
This hypothesis is supported by the observation that
CORT treatment potentiates the damaging effects of
kainic acid, an excitotoxin, on hippocampal CA3 pyramidal neurons in vivo31'39 and cultured hippocampal
neurons in vitro 33. In vivo, the damaging effects of
CORT in combination with kainic acid can be prevented
by administration of mannose 32. Presumably the presence
of an excess of this 'brain fuel' attenuates the synergy
between CORT and the neurotoxin. Both in vivo 16 and
in vitro 14 studies have indicated that glucocorticoids
decrease glucose uptake in hippocampal neurons and glia
providing a possible mechanism for CORT-induced
disruption of energy metabolism in the brain.
Direct effects of elevated levels of C O R T could be
mediated by either one or both of the corticosteroid
receptor systems in the rat brain. In the hippocampus,
CORT binds the type 1 receptor with high affinity and
the type 2 receptor with lower affinity25; type 1 receptors
are largely occupied by basal levels of endogenous
hormones whereas higher glucocorticoid levels are required to occupy the type 2 receptor25. As such, the
CORT-induced changes in dendritic morphology measured in this study are likely to be mediated by the type
2 receptor. It is, however, possible that the morphologic
effects we have observed are the result of changes in
receptor number which could involve either the type 1 or
the type 2 receptor system. Further study utilizing
selective receptor antagonists in vivo will be required to
distinguish between these possibilities.
Interestingly, we have observed CORT-induced
changes in dendritic morphology only in hippocampal
CA3 pyramidal cells. No differences in morphological
parameters were detected in CA1 pyramidal cells or
granule cells of the dentate gyrus. Thus, within the
hippocampus, "the effects of chronic C O R T treatment on
dendritic morphology appear to be specific to neurons of
the CA3 region. Although this is consistent with previous
findings that CA3 neurons are particularly sensitive to
high levels of glucocorticoids29, it is a puzzling result as
the CA3 region has been shown to contain lower levels
of CORT receptors than other subregions in the hippocampus. Immunocytochemical localization7"42 as well as
in situ hybridization 37'43 have been used to demonstrate
fewer type 2 (glucocorticoid) receptors and less receptor
mRNA in the CA3 region compared to the CA1 region
or the dentate gyrus. In addition, binding studies have
shown that the CA3 region contains fewer type 2
receptors than the dentate gyrus and fewer type 1
receptors than either the dentate gyrus or CA1 region 25.
In situ hybridization studies have shown that the CA3
230
region contains approximately the same amount of type
1 (mineralocorticoid) receptor mRNA as other subfields
in the hippocampus 1'43.
These observations suggest that CA3 pyramidal cells
are specialized in some way other than increased corticosteroid receptor number to be particularly sensitive to
high levels of CORT. There are at least two, not mutually
exclusive, possibilities which could explain the heightened vulnerability of these neurons. First, as several
studies have linked at least one form of depolarizationinduced cell death with increased Ca 2+ influx 4'13,
Sloviter35 has suggested that CA3 pyramidal neurons
could be relatively poorly protected from the damaging
effects of high levels of intracellular Ca 2÷ because they
contain low levels of Ca 2÷ binding proteins. In support of
this hypothesis, Sloviter has recently demonstrated a
positive correlation between the seizure-resistance of
neurons within the hippocampus and the presence of the
Ca 2÷ binding proteins calbindin-D28k or parvalbumin 35.
As CORT has been shown to positively regulate Ca 2÷
conductance in hippocampal pyramidal neurons 17, it
seems quite possible that the effects of prolonged
corticosterone exposure on neurons of the CA3 region
could be exacerbated by an inability to sequester potentially damaging Ca e÷.
Second, the effects of CORT on CA3 pyramidal cells
could be mediated at least in part, indirectly by a
CORT-sensitive afferent population. In that light, it is
interesting to note that the granule cells of the dentate
gyrus contain high levels of both type 1 and type 2 CORT
receptors 25 and project heavily to CA3 pyramidal cells
via the mossy fibers 3. Studies have demonstrated that
selective damage to CA3 pyramidal cells as a result of
intraventricular injections of kainic acid requires intact
mossy fiber input 2z. In addition, stimulation of the
perforant path selectively damages pyramidal cells of the
CA3 region apparently due in part to mossy fiber
activation 34. Thus, in two cases where CA3 pyramidal
cell damage is presumed to be due to hyperexcitation, it
appears that the damaging effects are mediated, at least
in part, through excitatory innervation from the dentate
gyrus. The granule cells of the dentate gyrus have been
shown to be highly dependent on CORT as short-term
adrenalectomy without hormone replacement (3 and 7
days) results in massive cell death in this hippocampal
subfield H. Thus, it seems possible that excess CORT
could exert a stimulatory effect on these neurons which
could, in turn, hyperactivate and promote damage to the
pyramidal cells of the CA3 region. Further experimentation will be necessary in order to determine the effects
of elevated CORT levels on granule cell activity.
As the intracellular events ultimately resulting in CA3
pyramidal cell damage are not yet understood, it is
unclear to what extent the damaging effects of increased
activity of excitatory afferents could be local, i.e. restricted to the region of the dendritic tree receiving input,
or generalized throughout the neuron. Interestingly, the
mossy fibers of the dentate gyrus granule cells make
synaptic contact with CA3 pyramidal cells primarily on
specialized spines located on the apical dendritic shaft 3's.
We have observed the effects of excess CORT only in the
apical dendritic tree; the basal dendritic tree, which is far
less densely innervated by the dentate gyrus 8, is apparently not morphologically sensitive to our CORT treatment. It is also interesting to note that we observed no
morphological effects of CORT treatment on CA1 pyramidal cells, a region to which the dentate gyrus has no
known projection in the rostral hippocampus of the rat 8.
The changes in dendritic morphology we have observed are likely to have profound consequences for
hippocampal neuronal function. Differences in dendritic
branching patterns ~5 and in amount of dendritic
material 24 have been suggested to modulate the spatial/
temporal integration of synaptic input to an individual
neuron. Decreases in these parameters could result in a
smaller total dendritic surface area available for synaptic
contact and/or could alter the pathway for the flow of
current in the dendrites. Thus, quantitatively less dramatic but qualitatively similar changes in hippocampal
dendritic morphology may be involved in the alterations
in hippocampal function which have been demonstrated
under physiological conditions in which CORT levels are
chronically elevated such as in aged 19"26~28'4° or chronically stressed animals eT.
In addition, if the apparent atrophy of the apical
dendritic tree in CA3 pyramidal cells does, in fact, reflect
neuronal degeneration in this region, then our study may
shed some light on morphological transitional phases
undergone by neurons in the process of degeneration. In
other words, the fact that we have observed decreases in
parameters of apical dendritic arborization with elevated
levels of glucocorticoids, i.e. under conditions which have
been suggested to result in prolonged catabolism 3°'32,
might indicate that this region of the neuron is energetically costly to maintain and that it may be compromised
in an unhealthy cell.
While the ultimate functional consequences of hippocampal exposure to elevated glucocorticoid levels are not
yet understood, it is clear that, in addition to the neuron
loss that has been reported by others 2°'29, there are
dramatic differences in the dendritic architecture of the
remaining CA3 pyramidal cells. Future studies will be
required in order to determine the extent to which Ca z÷
binding proteins and/or excitatory afferents play a role in
mediating the neuronal specificity of glucocorticoid effects on neuronal morphology.
231
REFERENCES
1 Arriza, J.L., Simerly, R.B., Swanson, L.W. and Evans, R.M.,
The neuronal mineralocorticoid receptor as a mediator of
glucocorticoid response, Neuron 1, (1988) 887-900.
2 Aus der Muhlen, K. and Ockenfels, H., Morphologische
Veranderungen im Diencephalon und Telencephalon Storungen
des Regalkreises Adenohypophyse-Nebennie.renrinde, Z. Zelllurch. Mikrosk. Anat., 93 (1969) 126-141.
3 Bayer, S.A., Hippocampal region. In G. Paxinos (Ed.), The Rat
Nervous System, Vol. 1, Academic, Sydney, 1985, pp. 335-352.
4 Choi, D.W., Ionic dependence of glutamate neurotoxicity, J.
Neurosci., 7 (1987) 369-379.
5 Feldman, S. and Conforti, N., Participation of the dorsal
hippocampus in the glucocorticoid feedback effect on adrenocortical activity, Neuroendocrinology, 30 (1980) 52-55.
6 Foy, M.R., Stanton, M,E., Levine, S. and Thompson, R.E,
Behavioral stress impairs long-term potentiation in rodent
hippocampus, Behav. and Neural Biol., 48 (1987) 138-149.
7 Fuxe, K., Wilkstrom, A., Okret, S., Agnati, L.E., Harfstrand, A.,
Yu, Z., Granholm, L., Zoli, M., Vale, W. and Gustafsson, J.,
Mapping of glucocorticoidreceptor immunoreactive neurons in the
rat tel- and diencephalon using a monoclonal antibody against rat
glucocorticoid receptor, Endocrinology, 117 (1985) 1803-1812.
8 Gaarskjaer, EB., The organization and development of the
hippocampal mossy fiber system, Brain Res. Rev., 11 (1986)
335-357.
9 Gabbott, EL. and Somogyi, J., The 'single' section Golgi
impregnation procedure: methodological description, J. Neurosci. Methods, 11 (1984) 221-230.
10 Gold, P.E. and McGaugh, J.L., Changes in learning and
memory during aging. In J.M. Ordy and K.R. Brizzee (Eds.),
Neurobiology of Aging, Plenum, New York, 1975.
11 Gould, E., Woolley, C.S. and McEwen, B.S., Short-term
glucocorticoid manipulations affect neuronal morphology and
survival in the adult dentate gyrus, Neuroscience, in press.
12 Hauger, R.L., Millam, M.A., Catt, K.J. and Aguilera, G.,
Differential regulation of brain and pituitary corticotropinreleasing factor receptors by corticosterone, Endocrinology, 120
(1987) 1527-1533.
13 Hori, N., Ffrench-Mulen, J.M.H. and Carpenter, D.O., Kainic
acid response and toxicity show pronounced Ca2+ dependence,
Brain Research, 175 (1985) 341-346.
14 Homer, H.C. and Sapolsky, R.M., Glucocorticoids decrease
glucose transport in cultured hippocampal neurons, Soc. Neurosci. Abstr., 14 (1988) 921.
15 Horwitz, B., Neuronal plasticity: how changes in dendritic
architecture can affect the spread of post-synaptic potentials,
Brain Research, 224 (1981) 412-418.
16 Kadekaro, M., Ito, M.M. and Gross, P.M., Local cerebral
glucose utilization is increased in acutely adrenalectomized rats,
Neuroendocrinology, 47 (1988) 329-334.
17 Kerr, D.S., Campbell, L.W., Hao, S. and Landfield, P.W.,
Corticosteroid modulation of hippocampal potentials: increased
effect with aging, Science, 245 (1989) 1505-1509.
18 Knigge, K. and Hays, M., Evidence of inhibitive role of
hippocampus in neural regulation of ACTH release, Proc. Soc.
Exp. Biol. Med., 114 (1963) 67-69.
19 Landfield, P.W., Waymire, J. and Lynch, G., Hippocampal
aging and adrenocorticoids: quantitative correlations, Science,
202 (1978) 1098-1102.
20 Landfield, P.W., Baskin, E. and Pitier, T., Brain aging correlates: retardation by hormonal-pharmacological treatments, Science, 214 (1981) 581-584.
21 Miller, M.M., Antecka, E. and Sapolsky, R.M., Short-term
effects of glucocorticoids on hippocampal ultrastructure, Exp.
Brain Res., 77 (1989) 309-314.
22 Nadler, J.V. and Cuthbertson, G.J., Kainic acid neurotoxicity
toward hippocampal formation: dependence on specific excitatory pathways, Brain Research, 195 (1980) 47-56.
23 Olton, D.D.S., Memory functions and the hippocampus. In W.
Seifert (Ed.), Neurobiology of the Hippocampus, Academic,
London, 1983, pp. 335-373.
24 Purves, D,, Regulation of neural connections in maturity. In
Body and Brain, ATrophic Theory of Neural Connections,
Harvard Univ. Press, Cambridge, 1988, pp. 97-122.
25 Reul, J.M.H. and De Kloet, E.R., Two receptor systems for
glucocorticoids in the rat brain: microdistribution and differential occupation, Endocrinology, 117 (1985) 2505-2511.
26 Sapolsky, R.M., Krey, L.C. and McEwen, B.S., The adrenocortical stress response in the aged male rat: impairment of
recovery from stress, Exp. Gerontol., 18 (1983) 55-64.
27 Sapolsky, R.M., Krey, L.C. and McEwen, B.S., Stress downregulates corticosterone receptors in a site-specific manner in the
rat brain, Endocrinology, 114 (1984 287-292.
28 Sapoisky, R.M., Krey, L.C. and McEwen, B.S., Glucocorticoidsensitive hippocampal neurons are involved in terminating the
~drenocortical stress response, Proc. Natl. Acad. Sci. U.S.A.. 84
(1984) 6174-6177.
29 Sapolsky, R.M., Krey, L.C. and McEwen, B.S., Prolonged
glucocorticoid exposure reduces hippocampal neuron number:
implications for aging, J. Neurosci., 5 (1985) 1222-1227.
30 Sapolsky, R.M., A mechanism for glucocorticoid toxicity in the
hippocampus: increased vulnerability to metabolic insults, J.
Neurosci., 5 (1985) 1228-1232.
31 Sapolsky, R.M., Glucocorticoid toxicity in the hippocampus:
temporal aspects of synergy with kainic acid, Neuroendocrinology, 40 (1986) 440-444.
32 Sapolsky, R.M., Glucocorticoid toxicity in the hippocampus:
reversal by supplementation with brain fuels, J. Neurosci., 6
(1986) 2240-2244.
33 Sapolsky, R.M., Packan, D.R. and Vale, W.W., Giucocorticoid
toxicity in the hippocampus: in vitro demonstration, Brain
Research, 473 (1988) 367-371,
34 Sloviter, R.S., 'Epileptic' brain damage in rats induced by
sustained electrical stimulation of the perforant path. I. Acute
electrophysiological and light microscopic studies, Brain Res.
Bull., 10 (1983) 675-697.
35 Sloviter, R.S., Calcium binding protein (calbindin-D28k) and
parvalbumin immunocytochemistry: localization in the rat hippocampus with specific reference to the selective vulnerability of
hippocampal neurons to seizure activity, J. Comp. Neurol., 280
(1989) 183-196.
36 Smith, G.M., Miller, R.H. and Silver, J., Changing 'role of
forebrain astrocytes during development, regenerative failure
and induced regeneration upon transplantation, J.-Comp.
Neurol., 251 (1986) 23-43.
37 Sousa, R.J., Tannery, N.H. and Lafer, E.M., In situ hybridization mapping of glucocorticoid receptor messenger ribonucleic
acid in the rat brain, Mol. Endocrinol., 3 (1989) 481-494.
38 Squire, L.R., The hippocampus and the neuropsychology of
memory. In W. Seifert (Ed.) Neurobiology of the Hippocampus,
Academic, London, 1983, pp. 491-511.
39 Stein, B.A. and Sapolsky, R.M., Chemical adrenalectomy
reduces hippocampal damage induced by kainic acid, Brain
Research, 473 (1988) 175-180.
40 Tang, G. and Phillips, R., Some age-related changes in pituitary
adrenal function in the male laboratory rat, J. Gerontol., 33
(1978) 377-382.
41 Uno, H., Tarara, R., Else, J.G., Suleman, M.A. and Sapolsky,
R.M., Hippocampal damage associated with prolonged and fatal
stress in primates, J. Neurosci., 9 (1989) 1705-1711.
42 Van Eekelen, J.A.M., Kiss, J.Z., Westphal, H.M. and De
Kloet, E.R., Immunocytochemical study on the intracellular
localization of the type 2 glucocorticoid receptor in the rat brain,
Brain Research, 436 (1987) 120-128.
43 Van Eekelen, J.A.M., Jiang, W., De Kloet, E.R. and Bohn,
M.C., Distribution of the mineralocorticoid and glucocorticoid
receptor mRNAs in the rat hippocampus, J. Neurosci. Res., 21
(1988) 88-94.